Advancing Creativity: A Comprehensive Review of AI-Driven Text-to-Image Generation and Its Applications | ||||
| Advanced Sciences and Technology Journal | ||||
| Volume 2, Issue 2, December 2025, Page 1-17 PDF (1.82 MB) | ||||
| Document Type: Review Article | ||||
| DOI: 10.21608/astj.2025.343418.1018 | ||||
 View on SCiNiTO | ||||
| Authors | ||||
Noha Hussen   ; Ahmed Samir; Aliaa Adel; Abdelrahman Gaber; Mommen Attaia; Ahmed Mohamed
| ||||
| Software Engineering Department, Faculty of Engineering and Technology, Egyptian Chinese University, Cairo, Egypt | ||||
| Abstract | ||||
| The field of AI-driven text-to-image generation has emerged as a transformative intersection of technology and creativity, enabling the automatic synthesis of visuals from textual descriptions. This capability has profound implications for diverse applications, from storytelling and education to digital art and design. By automating the translation of textual content into visually rich representations, text-to-image generation bridges the gap between linguistic and visual modalities, fostering novel opportunities for innovation and exploration. This review explores the state-of-the-art advancements in text-to-image synthesis, emphasizing the technological evolution from Generative Adversarial Networks (GANs) to diffusion models and transformer-based architectures. It highlights how these models, including tools like DALL-E-2, Midjourney, and Stable Diffusion, have advanced in generating semantically aligned, visually coherent, and aesthetically appealing images. Despite notable progress, significant challenges remain. These include maintaining contextual coherence across sequences, adhering to artistic and compositional principles, and addressing the dependency on detailed textual prompts. Moreover, the limitations of existing evaluation metrics, such as the Inception Score (IS) and Fréchet Inception Distance (FID), are critically analyzed, underscoring the need for metrics that account for semantic fidelity, emotional resonance, and user-centric perspectives. The review synthesizes insights from recent studies to identify key areas for innovation, such as enhanced context management, integration of 3D modeling capabilities, and real-time user interaction mechanisms. Finally, the paper outlines future directions to address current limitations, promote interdisciplinary collaboration, and establish ethical guidelines for responsible AI deployment. By doing so, this work aims to provide a comprehensive foundation for advancing generative AI and its applications across creative industries | ||||
| Keywords | ||||
| Book Illustration Context Transition Generative AI Narrative Coherence Prompt Engineering Text; Image Synthesis Text; to; Image Illustration User Experience | ||||
|
Statistics Article View: 471 PDF Download: 413 |
||||
